Existing systems for monitoring video signals from multiple sources either have relied upon switching means to sequence through each of the sources in a predetermined pattern, or else implement some form of "split screen" display, in which a conventional video monitor displays, for example, the output of four cameras, each occupying a portion of the display screen. Often, the video output is also supplied to signal storage means such as a VCR which is operated in a single-frame recording mode, providing a time-lapse record of the camera outputs. These types of systems are in common usage, and are well known in the art.
More recently, the availability of digital techniques for video signal processing and data compression has opened new video monitoring alternatives. Mathisen, for example, U.S. Pat. No. 4,198,656, describes a system employing a signal selector which is controlled by signal storage means, including a VCR functioning as a time-lapse frame recorder, and disc storage means functioning as a frame-store type device which provides continuity in the display while the selector is advancing through its sequence to the next signal source, or to the next recorded signal from one particular source. Each image is further provided with a digital code, which enables the image source to be identified upon playback.
Katz (U.S. Pat. No. 5,216,502) discloses a system for monitoring and recording transactions at multiple cashier lanes. According to the preferred embodiment, output signals from four cameras are fed to a four-quadrant multiplexer, which contains a frame store having provisions for reducing each camera image to one-quarter size and then displaying it in one of the four quadrants of the video monitor. The combined output signal is then recorded on a Video Cassette Recorder (VCR) to provide a more permanent record of the transaction. No description is given of the size-reduction method, but this kind of display may be implemented by employing a time-compression scheme for the horizontal lines, and eliminating alternate scan lines from the vertical dimension of the video frame. Data from the individual cash registers is encoded digitally, and then either recorded on lines of video in the vertical-interval or else recorded on the audio tracks of the VCR. The monitoring provisions include a video overlay generator, so that transaction information may be displayed concurrently with images of the event. Provisions are also included for selective recording of representative frames of video, as triggered by transaction events at individual cashier lanes. This further reduces the number of frames recorded, and extends the recording duration of the video cassette.
Neither of these two inventions disclose the use of image data compression schemes, nor of image resizing by digital means. Katz does disclose selective recording of frames of video, but only in analog form, and using conventional video recording means, such as a VCR.
Blum et al. (U.S. Pat. No. 5,237,408) disclose a system employing multiple interface circuit boards containing provisions for capturing and digitizing images from a plurality of cameras, further storing a slow-scan sequence of images for each camera in active buffer memory means as RAM on each of the interface boards. The display of these image sequences is preferably controlled by a PC with disk storage capabilities, and individual images selectively may be recorded on the hard disk, under operator control. Storage capacity is limited, however, because no image data compression is provided, and it is not practical to store large quantities or sequences of images. There are no provisions for automatic recording of images, nor any provision for storing these images in a data form differing from the display format image size.
Gormley (U.S. Pat. No. 5,258,837) discloses video compression means as a method for fitting multiple camera images within a single video display screen. However, Gormley clearly indicates that the system relies on fixed image-size compression of the digitized incoming video signal, and does not perform any kind of bandwidth compression. Provisions are included for storing images on a video recorder, but, as in other systems, these images are stored by taking the screen display as a whole, rather than storing independently constituted representations of the individual camera outputs.
In general, these conventional systems, regardless of the subsequent signal processing employed, accept analog video signals from the camera sources for input to the monitoring system. As such, these signals are susceptible to disturbances such as RF interference, ground-loop noise, and high-frequency signal loss due to long runs of coaxial cables. Automatic video switcher units sequence the source signals at a relatively slow rate, requiring as much as 16 seconds delay between successive viewings of a particular source when using a one source-per-second rate to sequence through 16 sources. When such a system is equipped with a time-lapse video recorder, the recording medium is expensive, has low resolution, and is limited in recording capacity.
Other systems, such as video conferencing arrangements, use multiple camera images and audio sources which are coordinated through electronic switching means and image display means, with interconnections between conferencing rooms locally or at remote sites via telephone lines or other communications links. Fabris et al. (U.S. Pat. No. 4,516,156) discloses a sophisticated implementation of such a system, employing satellite links for long-distance communications, a full range of camera remote controls (including lens controls and pan-tilt-head controls), and touch-screen monitor facilities for system controls of cameras and signal switching. This system, however, has no provisions for image storage, or any application of image data compression means coupled with display means or image storage means.